Dataset Overview

This dataset provides information on used car sales, including key variables that capture various attributes of the cars, such as their make, model, year of manufacture, transmission type, odometer reading, and selling price. The dataset also contains additional details about the condition of the cars, state of sale, and more.

Key columns in the dataset:

• Year: The year the car was manufactured.
• Make: The brand of the car.
• Model: The specific model of the car.
• Trim: A variant or version of the model with different features.
• Body: The type or category of the car (e.g., SUV, Sedan).
• Transmission: Type of transmission (automatic or manual).
• VIN: Vehicle Identification Number (unique identifier for each car).
• State: The state where the car is sold.
• Condition: The condition of the car when sold (e.g., new, used).
• Odometer: Number of miles the car has been driven.
• Color: The exterior color of the car.
• Interior: The interior color/material of the car.
• Seller: The seller of the car.
• MMR: Manheim Market Report value (provides a price benchmark for the vehicle).
• Selling Price: The actual selling price of the car.
• Sale Date: The date the car was sold.

Data Preparation

1. Importing Necessary Libraries

import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import plotly.express as px
import numpy as np
import warnings

warnings.filterwarnings("ignore")
df = pd.read_csv(r'car_prices.csv')
df.head()

	year	make	model	trim	body	transmission	vin	state	condition	odometer	color	interior	seller	mmr	sellingprice	saledate
0	2015	Kia	Sorento	LX	SUV	automatic	5xyktca69fg566472	ca	5.0	16639.0	white	black	kia motors america inc	20500.0	21500.0	Tue Dec 16 2014 12:30:00 GMT-0800 (PST)
1	2015	Kia	Sorento	LX	SUV	automatic	5xyktca69fg561319	ca	5.0	9393.0	white	beige	kia motors america inc	20800.0	21500.0	Tue Dec 16 2014 12:30:00 GMT-0800 (PST)
2	2014	BMW	3 Series	328i SULEV	Sedan	automatic	wba3c1c51ek116351	ca	45.0	1331.0	gray	black	financial services remarketing (lease)	31900.0	30000.0	Thu Jan 15 2015 04:30:00 GMT-0800 (PST)
3	2015	Volvo	S60	T5	Sedan	automatic	yv1612tb4f1310987	ca	41.0	14282.0	white	black	volvo na rep/world omni	27500.0	27750.0	Thu Jan 29 2015 04:30:00 GMT-0800 (PST)
4	2014	BMW	6 Series Gran Coupe	650i	Sedan	automatic	wba6b2c57ed129731	ca	43.0	2641.0	gray	black	financial services remarketing (lease)	66000.0	67000.0	Thu Dec 18 2014 12:30:00 GMT-0800 (PST)

2. Data Overview

df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 558837 entries, 0 to 558836
Data columns (total 16 columns):
 #   Column        Non-Null Count   Dtype  
---  ------        --------------   -----  
 0   year          558837 non-null  int64  
 1   make          548536 non-null  object 
 2   model         548438 non-null  object 
 3   trim          548186 non-null  object 
 4   body          545642 non-null  object 
 5   transmission  493485 non-null  object 
 6   vin           558833 non-null  object 
 7   state         558837 non-null  object 
 8   condition     547017 non-null  float64
 9   odometer      558743 non-null  float64
 10  color         558088 non-null  object 
 11  interior      558088 non-null  object 
 12  seller        558837 non-null  object 
 13  mmr           558799 non-null  float64
 14  sellingprice  558825 non-null  float64
 15  saledate      558825 non-null  object 
dtypes: float64(4), int64(1), object(11)
memory usage: 68.2+ MB

df.describe().T

	count	mean	std	min	25%	50%	75%	max
year	558837.0	2010.038927	3.966864	1982.0	2007.0	2012.0	2013.0	2015.0
condition	547017.0	30.672365	13.402832	1.0	23.0	35.0	42.0	49.0
odometer	558743.0	68320.017767	53398.542821	1.0	28371.0	52254.0	99109.0	999999.0
mmr	558799.0	13769.377495	9679.967174	25.0	7100.0	12250.0	18300.0	182000.0
sellingprice	558825.0	13611.358810	9749.501628	1.0	6900.0	12100.0	18200.0	230000.0

df.count()

year            558837
make            548536
model           548438
trim            548186
body            545642
transmission    493485
vin             558833
state           558837
condition       547017
odometer        558743
color           558088
interior        558088
seller          558837
mmr             558799
sellingprice    558825
saledate        558825
dtype: int64

df.shape

(558837, 16)

df.columns = df.columns.str.strip().str.lower()

3. Data Cleaning

plt.figure(figsize=(12, 6))
sns.heatmap(df.isnull(), cbar=False, cmap='viridis')
plt.show()

df.nunique()

year                34
make                96
model              973
trim              1963
body                87
transmission         4
vin             550297
state               64
condition           41
odometer        172278
color               46
interior            17
seller           14263
mmr               1101
sellingprice      1887
saledate          3766
dtype: int64

df.isnull().sum()/(df.shape[0])*100

year             0.000000
make             1.843292
model            1.860829
trim             1.905922
body             2.361154
transmission    11.694287
vin              0.000716
state            0.000000
condition        2.115107
odometer         0.016821
color            0.134028
interior         0.134028
seller           0.000000
mmr              0.006800
sellingprice     0.002147
saledate         0.002147
dtype: float64

fill_methods = {
    'make': 'mode',
    'model': 'mode',
    'trim': 'mode',
    'body': 'mode',
    'transmission': 'mode',
}
for column, method in fill_methods.items():
    if method == 'median':
        df[column].fillna(df[column].median(), inplace=True)
    elif method == 'mode':
        df[column].fillna(df[column].mode()[0], inplace=True)
df.isnull().mean() * 100

columns_to_drop = ['vin']

df.drop(columns=columns_to_drop, inplace=True)

df.columns

Index(['year', 'make', 'model', 'trim', 'body', 'transmission', 'state',
       'condition', 'odometer', 'color', 'interior', 'seller', 'mmr',
       'sellingprice', 'saledate'],
      dtype='object')

df = df.dropna(subset=['sellingprice', 'saledate', 'mmr', 'make', 'color', 'odometer', 'condition', 'interior'])

df.isnull().mean() * 100

year            0.0
make            0.0
model           0.0
trim            0.0
body            0.0
transmission    0.0
state           0.0
condition       0.0
odometer        0.0
color           0.0
interior        0.0
seller          0.0
mmr             0.0
sellingprice    0.0
saledate        0.0
dtype: float64

df.shape

(546325, 15)

df.describe().T

	count	mean	std	min	25%	50%	75%	max
year	546325.0	2010.132280	3.906608	1982.0	2008.0	2012.0	2013.0	2015.0
condition	546325.0	30.669557	13.400986	1.0	23.0	35.0	42.0	49.0
odometer	546325.0	67289.743418	52781.847842	1.0	28072.0	51260.0	97438.0	999999.0
mmr	546325.0	13903.192697	9652.965754	25.0	7375.0	12400.0	18400.0	182000.0
sellingprice	546325.0	13751.715761	9724.318611	1.0	7100.0	12300.0	18300.0	230000.0

df.head()

	year	make	model	trim	body	transmission	state	condition	odometer	color	interior	seller	mmr	sellingprice	saledate
0	2015	Kia	Sorento	LX	SUV	automatic	ca	5.0	16639.0	white	black	kia motors america inc	20500.0	21500.0	Tue Dec 16 2014 12:30:00 GMT-0800 (PST)
1	2015	Kia	Sorento	LX	SUV	automatic	ca	5.0	9393.0	white	beige	kia motors america inc	20800.0	21500.0	Tue Dec 16 2014 12:30:00 GMT-0800 (PST)
2	2014	BMW	3 Series	328i SULEV	Sedan	automatic	ca	45.0	1331.0	gray	black	financial services remarketing (lease)	31900.0	30000.0	Thu Jan 15 2015 04:30:00 GMT-0800 (PST)
3	2015	Volvo	S60	T5	Sedan	automatic	ca	41.0	14282.0	white	black	volvo na rep/world omni	27500.0	27750.0	Thu Jan 29 2015 04:30:00 GMT-0800 (PST)
4	2014	BMW	6 Series Gran Coupe	650i	Sedan	automatic	ca	43.0	2641.0	gray	black	financial services remarketing (lease)	66000.0	67000.0	Thu Dec 18 2014 12:30:00 GMT-0800 (PST)

df.duplicated().sum()

0

4. Feature Engineering

Handling Date Data

df.head()
df['saledate'] = pd.to_datetime(df['saledate'], utc=True)
sales_per_day = df['saledate'].dt.date.value_counts()

top_10_sales_days = sales_per_day.nlargest(10)

print("Top 10 sales days:")
display(top_10_sales_days)

Top 10 sales days:

saledate
2015-01-13    20898
2015-02-10    15902
2015-06-02    13882
2015-02-24    13799
2015-02-03    13787
2015-02-17    13540
2015-02-09    13040
2015-01-20    12206
2015-03-03    12102
2015-06-16    12087
Name: count, dtype: int64

df['saledate'] = pd.to_datetime(df['saledate'], utc=True)
df['sale_year'] = df['saledate'].dt.year
df['sale_month'] = df['saledate'].dt.month
df['sale_day'] = df['saledate'].dt.day
df['sale_dayofweek'] = df['saledate'].dt.dayofweek
df.dtypes

year                            int64
make                           object
model                          object
trim                           object
body                           object
transmission                   object
state                          object
condition                     float64
odometer                      float64
color                          object
interior                       object
seller                         object
mmr                           float64
sellingprice                  float64
saledate          datetime64[ns, UTC]
sale_year                       int32
sale_month                      int32
sale_day                        int32
sale_dayofweek                  int32
dtype: object

Standardizing the condition Values

unique_conditions = df['condition'].unique()
unique_conditions

array([ 5., 45., 41., 43.,  1., 34.,  2., 42.,  3., 48., 49., 17., 19.,
       29., 38., 44., 47., 32.,  4., 25., 37., 39., 31., 28., 46., 36.,
       35., 26., 21., 22., 27., 24., 33., 23., 15., 16., 18., 12., 14.,
       11., 13.])

sns.lineplot(data=df, x='condition', y='sellingprice', color='blue')

plt.title('Selling Price by Condition', fontsize=16)
plt.xlabel('Condition', fontsize=12)
plt.ylabel('Selling Price ($)', fontsize=12)

plt.xticks(fontsize=10)
plt.yticks(fontsize=10)

plt.grid(False)

plt.tight_layout()
plt.show()

We will notice that the scale from 1 to 5 is the same as the scale from 10 to 50 and the same relationship, but narrower

def unify_condition(value):
    if 1 <= value <= 5:
        return value
    elif 10 <= value <= 50:
        return round(value / 10)
    else:
        return None

df['condition'] = df['condition'].apply(unify_condition)
df['condition'].unique()

array([5., 4., 1., 3., 2.])

df['condition'].isnull().sum()/(df.shape[0])*100

0.0

df['condition'] = df['condition'].astype(int)
record_counts = df['condition'].value_counts().reset_index()
record_counts

	condition	count
0	4	244240
1	3	124265
2	2	109184
3	5	60894
4	1	7742

Renaming States

state_names = {
    'ca': 'California', 'tx': 'Texas', 'pa': 'Pennsylvania', 'mn': 'Minnesota',
    'az': 'Arizona', 'wi': 'Wisconsin', 'tn': 'Tennessee', 'md': 'Maryland',
    'fl': 'Florida', 'ne': 'Nebraska', 'nj': 'New Jersey', 'nv': 'Nevada',
    'oh': 'Ohio', 'mi': 'Michigan', 'ga': 'Georgia', 'va': 'Virginia',
    'sc': 'South Carolina', 'nc': 'North Carolina', 'in': 'Indiana',
    'il': 'Illinois', 'co': 'Colorado', 'ut': 'Utah', 'mo': 'Missouri',
    'ny': 'New York', 'ma': 'Massachusetts', 'pr': 'Puerto Rico', 'or': 'Oregon',
    'la': 'Louisiana', 'wa': 'Washington', 'hi': 'Hawaii', 'qc': 'Quebec',
    'ab': 'Alberta', 'on': 'Ontario', 'ok': 'Oklahoma', 'ms': 'Mississippi',
    'nm': 'New Mexico', 'al': 'Alabama', 'ns': 'Nova Scotia'
}
df['state'] = df['state'].map(state_names)
df.head()

	year	make	model	trim	body	transmission	state	condition	odometer	color	interior	seller	mmr	sellingprice	saledate	sale_year	sale_month	sale_day	sale_dayofweek
0	2015	Kia	Sorento	LX	SUV	automatic	California	5	16639.0	white	black	kia motors america inc	20500.0	21500.0	2014-12-16 04:30:00+00:00	2014	12	16	1
1	2015	Kia	Sorento	LX	SUV	automatic	California	5	9393.0	white	beige	kia motors america inc	20800.0	21500.0	2014-12-16 04:30:00+00:00	2014	12	16	1
2	2014	BMW	3 Series	328i SULEV	Sedan	automatic	California	4	1331.0	gray	black	financial services remarketing (lease)	31900.0	30000.0	2015-01-14 20:30:00+00:00	2015	1	14	2
3	2015	Volvo	S60	T5	Sedan	automatic	California	4	14282.0	white	black	volvo na rep/world omni	27500.0	27750.0	2015-01-28 20:30:00+00:00	2015	1	28	2
4	2014	BMW	6 Series Gran Coupe	650i	Sedan	automatic	California	4	2641.0	gray	black	financial services remarketing (lease)	66000.0	67000.0	2014-12-18 04:30:00+00:00	2014	12	18	3

5. Outlier Handling

Removing Illogical Values from sellingprice

df['Price_diff']=df['sellingprice'] - df['mmr']

def outliers_handling(x, p1=0.25, p2=0.75):
    q1 = x.quantile(p1)
    q3 = x.quantile(p2)
    iqr = q3 - q1
    min_limit = q1 - 1.5 * iqr
    max_limit = q3 + 1.5 * iqr
    return x[(x > min_limit) & (x < max_limit)]

for col in ['odometer','mmr','sellingprice']:
    df[col] = outliers_handling(df[col])

df['Price_diff'] = outliers_handling(df['Price_diff'])

df.isnull().sum()/(df.shape[0])*100

year              0.000000
make              0.000000
model             0.000000
trim              0.000000
body              0.000000
transmission      0.000000
state             0.000000
condition         0.000000
odometer          1.959456
color             0.000000
interior          0.000000
seller            0.000000
mmr               3.016886
sellingprice      2.970759
saledate          0.000000
sale_year         0.000000
sale_month        0.000000
sale_day          0.000000
sale_dayofweek    0.000000
Price_diff        6.290395
dtype: float64

Removing the illogical differences between MMR and selling price in some rows

df['sell_price']=df['Price_diff']+ df['mmr']

Handling Missing and Inconsistent Values in Categorical Columns

Categorical_col = df.select_dtypes(include= 'object').columns

df[Categorical_col] = df[Categorical_col].apply(lambda col: col.str.upper())

df['seller'].unique()

array(['KIA MOTORS AMERICA  INC',
       'FINANCIAL SERVICES REMARKETING (LEASE)',
       'VOLVO NA REP/WORLD OMNI', ..., 'MASERATI NORTH AMERICA INC',
       'ALTERNATIVE FINANCIAL GROUP INC', 'I -5 UHLMANN RV'], dtype=object)

df['seller'] = df['seller'].str.lower().str.strip()
df['seller'] = df['seller'].str.replace('-', ' ', regex=False)

df['seller'].value_counts()

seller
nissan infiniti lt               29631
ford motor credit company llc    19103
the hertz corporation            18286
santander consumer               15277
avis corporation                 12510
                                 ...  
autotrust inc                        1
thoroughbred motors llc              1
titlemax/alexandria va4              1
fmi enterprises                      1
i  5 uhlmann rv                      1
Name: count, Length: 12951, dtype: int64

df['seller'].unique()

array(['kia motors america  inc',
       'financial services remarketing (lease)',
       'volvo na rep/world omni', ..., 'maserati north america inc',
       'alternative financial group inc', 'i  5 uhlmann rv'], dtype=object)

columns_with_weird_values = ['color', 'interior']
for column in columns_with_weird_values:
    weird_rows_count = df[df[column] == '—'].shape[0]
    total_rows = df.shape[0]
    percentage_weird_rows = (weird_rows_count / total_rows) * 100

    print(f'Column: {column}')
    print(f'Weird rows count: {weird_rows_count}')
    print(f'Percentage of weird rows: {percentage_weird_rows:.2f}%\n')

Column: color
Weird rows count: 24614
Percentage of weird rows: 4.51%

Column: interior
Weird rows count: 11844
Percentage of weird rows: 2.17%

for col in ['color', 'interior']:
    most_common_value = df[col].mode()[0]
    df[col] = df[col].replace('—', most_common_value)
    weird_rows_count = df[df[col] == '—'].shape[0]
    percentage_weird_rows = (weird_rows_count / df.shape[0]) * 100
    print(f"Percentage of rows with weird '{col}' values after replacement: {percentage_weird_rows:.2f}%")

Percentage of rows with weird 'color' values after replacement: 0.00%
Percentage of rows with weird 'interior' values after replacement: 0.00%

df.isnull().mean() * 100

year              0.000000
make              0.000000
model             0.000000
trim              0.000000
body              0.000000
transmission      0.000000
state             0.000000
condition         0.000000
odometer          1.959456
color             0.000000
interior          0.000000
seller            0.000000
mmr               3.016886
sellingprice      2.970759
saledate          0.000000
sale_year         0.000000
sale_month        0.000000
sale_day          0.000000
sale_dayofweek    0.000000
Price_diff        6.290395
sell_price        8.659864
dtype: float64

df = df.dropna(subset=['sell_price'])
df = df.dropna(subset=['odometer'])
df = df.drop(columns=['Price_diff'])
df = df.drop(columns=['sellingprice'])

df.shape

(488667, 19)

df.dtypes

year                            int64
make                           object
model                          object
trim                           object
body                           object
transmission                   object
state                          object
condition                       int32
odometer                      float64
color                          object
interior                       object
seller                         object
mmr                           float64
saledate          datetime64[ns, UTC]
sale_year                       int32
sale_month                      int32
sale_day                        int32
sale_dayofweek                  int32
sell_price                    float64
dtype: object

df.columns

Index(['year', 'make', 'model', 'trim', 'body', 'transmission', 'state',
       'condition', 'odometer', 'color', 'interior', 'seller', 'mmr',
       'saledate', 'sale_year', 'sale_month', 'sale_day', 'sale_dayofweek',
       'sell_price'],
      dtype='object')

df.head()

	year	make	model	trim	body	transmission	state	condition	odometer	color	interior	seller	mmr	saledate	sale_year	sale_month	sale_day	sale_dayofweek	sell_price
0	2015	KIA	SORENTO	LX	SUV	AUTOMATIC	CALIFORNIA	5	16639.0	WHITE	BLACK	kia motors america inc	20500.0	2014-12-16 04:30:00+00:00	2014	12	16	1	21500.0
1	2015	KIA	SORENTO	LX	SUV	AUTOMATIC	CALIFORNIA	5	9393.0	WHITE	BEIGE	kia motors america inc	20800.0	2014-12-16 04:30:00+00:00	2014	12	16	1	21500.0
2	2014	BMW	3 SERIES	328I SULEV	SEDAN	AUTOMATIC	CALIFORNIA	4	1331.0	GRAY	BLACK	financial services remarketing (lease)	31900.0	2015-01-14 20:30:00+00:00	2015	1	14	2	30000.0
3	2015	VOLVO	S60	T5	SEDAN	AUTOMATIC	CALIFORNIA	4	14282.0	WHITE	BLACK	volvo na rep/world omni	27500.0	2015-01-28 20:30:00+00:00	2015	1	28	2	27750.0
7	2014	CHEVROLET	CRUZE	1LT	SEDAN	AUTOMATIC	CALIFORNIA	2	28617.0	BLACK	BLACK	enterprise vehicle exchange / tra / rental / t...	11900.0	2014-12-16 05:00:00+00:00	2014	12	16	1	9800.0

df.describe().T

	count	mean	std	min	25%	50%	75%	max
year	488667.0	2010.157815	3.807584	1982.0	2008.0	2012.0	2013.0	2015.0
condition	488667.0	3.467335	0.943698	1.0	3.0	4.0	4.0	5.0
odometer	488667.0	65735.333892	46297.634564	1.0	29138.0	52298.0	96921.0	201485.0
mmr	488667.0	12890.621272	7471.288043	25.0	7300.0	12100.0	17500.0	34900.0
sale_year	488667.0	2014.923586	0.265660	2014.0	2015.0	2015.0	2015.0	2015.0
sale_month	488667.0	3.613115	3.027329	1.0	1.0	2.0	6.0	12.0
sale_day	488667.0	14.544819	8.660383	1.0	7.0	15.0	21.0	31.0
sale_dayofweek	488667.0	1.446928	1.240843	0.0	1.0	1.0	2.0	6.0
sell_price	488667.0	12839.207211	7561.022752	100.0	7200.0	12100.0	17500.0	37700.0

cols = df.columns.tolist()
cols.remove('sell_price')
cols.insert(cols.index('mmr') + 1, 'sell_price')
df = df[cols]

df.dtypes

year                            int64
make                           object
model                          object
trim                           object
body                           object
transmission                   object
state                          object
condition                       int32
odometer                      float64
color                          object
interior                       object
seller                         object
mmr                           float64
sell_price                    float64
saledate          datetime64[ns, UTC]
sale_year                       int32
sale_month                      int32
sale_day                        int32
sale_dayofweek                  int32
dtype: object

df.isnull().mean() * 100

year              0.0
make              0.0
model             0.0
trim              0.0
body              0.0
transmission      0.0
state             0.0
condition         0.0
odometer          0.0
color             0.0
interior          0.0
seller            0.0
mmr               0.0
sell_price        0.0
saledate          0.0
sale_year         0.0
sale_month        0.0
sale_day          0.0
sale_dayofweek    0.0
dtype: float64

to CSV for ML

df.to_csv('cleaned_car_prices.csv', index=False)

Exploratory Data Analysis

Univariant Analysis

Distributions

fig, axes = plt.subplots(1, 2, figsize=(14, 6))

sns.histplot(df['sell_price'], kde=True, ax=axes[0], color='blue')
axes[0].set_title('Distribution of Selling Price', fontsize=16)
axes[0].set_xlabel('Selling Price', fontsize=14)
axes[0].set_ylabel('Frequency', fontsize=14)
axes[0].grid(True)
sns.histplot(df['odometer'], kde=True, ax=axes[1], color='green')
axes[1].set_title('Distribution of Odometer Readings', fontsize=16)
axes[1].set_xlabel('Odometer (Miles)', fontsize=14)
axes[1].set_ylabel('Frequency', fontsize=14)
axes[1].grid(True)

plt.tight_layout()

plt.show()

plt.figure(figsize=(12, 6))
sns.countplot(x='condition', data=df, palette='viridis', edgecolor='black')  # استخدام palette لإضافة ألوان جذابة
plt.title('Distribution of Car Condition', fontsize=16)
plt.xlabel('Condition', fontsize=14)
plt.ylabel('Frequency', fontsize=14)
plt.grid(axis='y', linestyle='--', alpha=0.7)
plt.tight_layout()
plt.show()

plt.figure(figsize=(12, 6))
sns.set_theme(style="whitegrid")
sns.countplot(y='model', data=df, order=df['model'].value_counts().index[:10], palette='coolwarm')
plt.title('Top 10 Car Models', fontsize=16, fontweight='bold')
plt.xlabel('Count', fontsize=14)
plt.ylabel('Model', fontsize=14)
plt.xticks(fontsize=12)
plt.yticks(fontsize=12)
plt.tight_layout()
plt.show()

plt.figure(figsize=(12, 6))
sns.set_theme(style="whitegrid")
sns.countplot(y='trim', data=df, order=df['trim'].value_counts().index[:10], palette='Set2')
plt.title('Top 10 Car Trims', fontsize=16, fontweight='bold')
plt.xlabel('Count', fontsize=14)
plt.ylabel('Trim', fontsize=14)
plt.xticks(fontsize=12)
plt.yticks(fontsize=12)
plt.tight_layout()
plt.show()

plt.figure(figsize=(12, 6))
sns.set_theme(style="whitegrid")
sns.countplot(y='body', data=df, order=df['body'].value_counts().index[:10], palette='viridis')
plt.title('Top 10 Car Body Types', fontsize=16, fontweight='bold')
plt.xlabel('Count', fontsize=14)
plt.ylabel('Body Type', fontsize=14)
plt.xticks(fontsize=12)
plt.yticks(fontsize=12)
plt.tight_layout()
plt.show()

top_3_trims = df['trim'].value_counts().nlargest(3)
plt.figure(figsize=(8, 6))
ax = sns.barplot(x=top_3_trims.index, y=top_3_trims.values, palette="Blues_d", width=0.5)
ax.bar_label(ax.containers[0], fontsize=12, fontweight='bold')
plt.title('Top 3 Trim Sales', fontsize=18, fontweight='bold', color='darkblue')
plt.xlabel('Trim', fontsize=14, fontweight='bold')
plt.ylabel('Sale Amount', fontsize=14, fontweight='bold')
plt.xticks(fontsize=12)
plt.yticks(fontsize=12)
plt.grid(True, linestyle='--', alpha=0.6)
plt.tight_layout()
plt.show()

*Some Questions And Insights:*

What is the shape of cars on the market?

top_5_body = df['body'].value_counts().head(5)
plt.figure(figsize=(10, 6))
sns.barplot(x=top_5_body.index, y=top_5_body.values, palette='viridis')
plt.title('Top 5 Most Common Body Types', fontsize=16)
plt.xlabel('Body Type', fontsize=14)
plt.ylabel('Count', fontsize=14)
plt.xticks(rotation=45)

plt.tight_layout()
plt.show()

The distribution of car bodies indicates a strong preference for sedans and SUVs in the market.

What are the days and months in which the sale is most frequent?

fig, axs = plt.subplots(1, 2, figsize=(15, 6))
# sns.set(style="whitegrid")
sns.countplot(x='sale_month', data=df, ax=axs[0], palette='viridis')
axs[0].set_title('Number of Sales by Month', fontsize=16, fontweight='bold')
axs[0].set_xlabel('Month', fontsize=14)
axs[0].set_ylabel('Number of Sales', fontsize=14)
axs[0].tick_params(axis='x', labelsize=12)
axs[0].tick_params(axis='y', labelsize=12)
sns.countplot(x='sale_dayofweek', data=df, ax=axs[1], palette='plasma')
axs[1].set_title('Number of Sales by Day of Week', fontsize=16, fontweight='bold')
axs[1].set_xlabel('Day of Week', fontsize=14)
axs[1].set_ylabel('Number of Sales', fontsize=14)
axs[1].tick_params(axis='x', labelsize=12)
axs[1].tick_params(axis='y', labelsize=12)
plt.tight_layout()
plt.show()

Tuesday is the most peak day for sale & February is the best-selling month in the data

What year is the best-selling car?

plt.figure(figsize=(15, 6))
sns.countplot(x='year', data=df, palette='viridis')

plt.title('Number of Cars Sold by Year', fontsize=18, fontweight='bold')
plt.xticks(rotation=45, ha='right', fontsize=12)
plt.xlabel('Year', fontsize=14)
plt.ylabel('Number of Sales', fontsize=14)

plt.grid(axis='y', linestyle='--', alpha=0.7)
plt.tight_layout()
plt.show()

2012 cars are the best sellers

What are the best-selling brands on the market?

top_makes = df['make'].value_counts().head(10)

plt.figure(figsize=(12, 7))
sns.barplot(x=top_makes.index, y=top_makes.values, palette='viridis')
plt.title('Top 10 Car Makes by Sales Count', fontsize=16, fontweight='bold')
plt.xlabel('Car Make', fontsize=14)
plt.ylabel('Number of Cars Sold', fontsize=14)

plt.xticks(rotation=45, ha='right', fontsize=12)
plt.yticks(fontsize=12)
plt.grid(axis='y', linestyle='--', alpha=0.7)

plt.tight_layout()
plt.show()

Ford, Chevrolet then nissan are the most in demand

What is the average price in the market?

plt.figure(figsize=(12, 6))
sns.histplot(df['sell_price'], bins=30, kde=True, color='lightblue', edgecolor='black')
plt.title('Distribution of Selling Price', fontsize=16)
plt.xlabel('Selling Price', fontsize=14)
plt.ylabel('Frequency', fontsize=14)
plt.grid(axis='y', linestyle='--', alpha=0.7)
plt.tight_layout()
plt.show()

We will notice that most of the prices are in the range of 10000 and 15000

Are most cars on the market automatic?

sns.set_style('whitegrid')

fig = plt.figure(figsize=(12, 6))
axes = fig.subplots(1, 2)

transmission = df['transmission'].value_counts()

axes[0].pie(transmission,
             labels=transmission.index,
             autopct="%1.1f%%",
             explode=[0, 0.1],
             colors=['#00bfff', '#ff4d4d'],
             startangle=140)
axes[0].set_title('Transmission Percentage', fontsize=16)

sns.countplot(data=df, x='transmission', ax=axes[1], palette='viridis')
axes[1].set_title('Transmission Count', fontsize=16)
axes[1].set_xlabel('Transmission', fontsize=14)
axes[1].set_ylabel('Count', fontsize=14)

plt.tight_layout()
plt.show()

96.9% Automatic and 3.1% Manual

What colors do buyers prerfer?

plt.figure(figsize=(12, 6))
sns.set(style="whitegrid")

custom_palette = {
    'BLACK': 'black',
    'WHITE': 'white',
    'SILVER': 'silver',
    'GRAY': 'gray',
    'BLUE': 'blue',
    'RED': 'red',
    'GOLD': 'gold',
    'GREEN': 'green',
    'BEIGE': 'beige',
    'BURGUNDY': 'maroon',
    'BROWN': 'brown',
    'ORANGE': 'orange',
    'PURPLE': 'purple',
    'OFF-WHITE': 'lightgray',
    'YELLOW': 'yellow',
    'CHARCOAL': 'dimgray',
    'TURQUOISE': 'turquoise',
    'PINK': 'pink',
    'LIME': 'lime'
}

color_counts = df['color'].value_counts().reset_index()
color_counts.columns = ['color', 'count']

bar_plot = sns.barplot(x='color', y='count', data=color_counts, palette=custom_palette)
plt.xticks(rotation=45, ha='right')

for bar in bar_plot.patches:
    bar.set_edgecolor('black')
    bar.set_linewidth(1.5)
plt.title('Count of Cars by Color', fontsize=16)
plt.xlabel('Color', fontsize=14)
plt.ylabel('Count', fontsize=14)
plt.tight_layout()
plt.show()

Black color is the most requested as expected

Where are the most sold cars?

state_abbreviations = {
    'CALIFORNIA': 'CA',
    'NEW JERSEY': 'NJ',
    'GEORGIA': 'GA',
    'VIRGINIA': 'VA',
    'INDIANA': 'IN',
    'ILLINOIS': 'IL',
    'MINNESOTA': 'MN',
    'MICHIGAN': 'MI',
    'OHIO': 'OH',
    'TEXAS': 'TX',
    'ARIZONA': 'AZ',
    'COLORADO': 'CO',
    'MISSOURI': 'MO',
    'PENNSYLVANIA': 'PA',
    'NEBRASKA': 'NE',
    'NEVADA': 'NV',
    'MASSACHUSETTS': 'MA',
    'UTAH': 'UT',
    'PUERTO RICO': 'PR',
    'NORTH CAROLINA': 'NC',
    'FLORIDA': 'FL',
    'SOUTH CAROLINA': 'SC',
    'NEW YORK': 'NY',
    'WISCONSIN': 'WI',
    'MARYLAND': 'MD',
    'TENNESSEE': 'TN',
    'WASHINGTON': 'WA',
    'LOUISIANA': 'LA',
    'OREGON': 'OR',
    'HAWAII': 'HI',
    'OKLAHOMA': 'OK',
    'MISSISSIPPI': 'MS',
    'NEW MEXICO': 'NM',
    'ALABAMA': 'AL'
}
df_temp = df['state'].replace(state_abbreviations)

state_counts = df_temp.value_counts().reset_index()
state_counts.columns = ['state', 'count']

fig = px.choropleth(
    state_counts,
    locations='state',
    locationmode='USA-states',
    color='count',
    scope="usa",
    color_continuous_scale="Blues",
    title="Distribution of Selling Price Across States"
)

fig.show()

Most sales are made in Florida and then California.

Bivariante Analysis

Some Visualizations to understand the data in a better way

newdata = df[['sell_price', 'odometer']]
newdata['odometer_group'] = (newdata['odometer'] // 5000) * 5000
avg_price_by_group = newdata.groupby('odometer_group')['sell_price'].mean().reset_index()

avg_price_by_mileage = df.groupby('odometer')['sell_price'].mean().reset_index()

fig, axs = plt.subplots(1, 2, figsize=(18, 8))

sns.lineplot(data=avg_price_by_group, x='odometer_group', y='sell_price', marker='o', color='green', ax=axs[0])
axs[0].set_title('Average Selling Price vs Odometer (Rounded to Nearest 5000 Miles)', fontsize=16)
axs[0].set_xlabel('Odometer (Miles)', fontsize=14)
axs[0].set_ylabel('Average Selling Price ($)', fontsize=14)
axs[0].tick_params(axis='x', rotation=45, labelsize=10)

sns.scatterplot(data=avg_price_by_mileage, x='odometer', y='sell_price', ax=axs[1])
axs[1].set_title('Average Selling Price by odometer', fontsize=16)
axs[1].set_xlabel('Odometer (Miles)', fontsize=14)
axs[1].set_ylabel('Average Selling Price ($)', fontsize=14)

plt.tight_layout()
plt.show()

The more distances a car travels, the less its value is in the market

plt.figure(figsize=(10, 6))
sns.lineplot(x='mmr', y='sell_price', data=df)
plt.title('Selling Price vs MMR')
plt.xlabel('MMR')
plt.ylabel('Selling Price')
plt.show()

The selling price and the MMR are approximately equal

avg_price_by_condition = df.groupby('condition')['sell_price'].mean().reset_index()

plt.figure(figsize=(14, 12))

plt.subplot(2, 1, 1)  
sns.lineplot(data=avg_price_by_condition, x='condition', y='sell_price', marker='o', color='royalblue')
plt.title('Average Selling Price by Condition', fontsize=18, fontweight='bold')
plt.xlabel('Condition', fontsize=14)
plt.ylabel('Average Selling Price ($)', fontsize=14)
plt.xticks(fontsize=12)
plt.yticks(fontsize=12)
plt.grid(True, linestyle='--', alpha=0.7)

df['condition_group'] = pd.cut(df['condition'],
                               bins=[0, 1, 2, 3, 4, 5],
                               labels=['Very Poor', 'Poor', 'Average', 'Good', 'Excellent'],
                               include_lowest=True)

plt.subplot(2, 1, 2)  
sns.boxplot(data=df, x='condition_group', y='sell_price', palette='Set2', 
            order=['Very Poor', 'Poor', 'Average', 'Good', 'Excellent'])
plt.title('Box Plot of Selling Price by Condition Group', fontsize=18, fontweight='bold')
plt.xlabel('Condition Group', fontsize=14)
plt.ylabel('Selling Price ($)', fontsize=14)
plt.xticks(rotation=45, fontsize=12)
plt.yticks(fontsize=12)
plt.grid(True, linestyle='--', alpha=0.7)

plt.tight_layout()
plt.show()

The average selling price of cars significantly increases with better condition ratings. Notably, cars rated as "Excellent" command the highest prices, while those rated as "Very Poor" show a wide price range, indicating potential losses for sellers.

plt.figure(figsize=(12, 6))  
sns.lineplot(data=df, x='year', y='odometer', color='green', marker='o')  

plt.title('Year vs. Odometer', fontsize=18, fontweight='bold')  
plt.xlabel('Year', fontsize=14)  
plt.ylabel('Odometer (Miles)', fontsize=14)  
plt.xticks(fontsize=12)  
plt.yticks(fontsize=12)  
plt.grid(True, linestyle='--', alpha=0.7)  

plt.tight_layout()  
plt.show()

Mostly the newer the car, the less the distance traveled than the old one

plt.figure(figsize=(10, 6))

sns.lineplot(x='year', y='condition', data=df, marker='o', linestyle='--', color='royalblue')

plt.title('Average Condition of Sold Cars Over the Years', fontsize=16, fontweight='bold')
plt.xlabel('Year', fontsize=14)
plt.ylabel('Condition of Sold Car', fontsize=14)
plt.ylim(1, 5)
plt.xticks(fontsize=12)
plt.yticks(fontsize=12)

plt.grid(True, linestyle='--', alpha=0.6)

plt.tight_layout()

plt.show()

The older cars are in worse condition than the new one because the situation worsens with the passage of time

Analysis of Car Sales and Total Revenue Over Years

import plotly.graph_objects as go

p = df.groupby('year').agg({'sell_price': ['count', 'sum']})
p.columns = ['sales_count', 'total_revenue']

fig = go.Figure()
fig.update_layout(yaxis=dict(title='Car Sold'), yaxis2=dict(title='Total Revenue', overlaying='y', side='right'), title='Distribution of Car Year and Total Revenue', template='plotly_white')
trace_1 = go.Bar(x=p.index, y=p['sales_count'])
trace_2 = go.Scatter(x=p.index, y=p['total_revenue'], yaxis='y2', mode='lines+markers', line=dict(color='green'))
fig.add_traces([trace_1, trace_2])

fig.update_traces(selector=dict(type='bar'), name='Sales Count')
fig.update_traces(selector=dict(type='scatter'), name='Total Revenue')
fig.show()

What is the average car prices in the market?

def categorize(price):
    if price < 10000:
        return 'Economical'
    elif 10000 <= price < 20000:
        return 'Medium'
    else:
        return 'High'

# Add a category column
df['category'] = df['sell_price'].apply(categorize)

# Group by make and calculate the average price
df_avg = df.groupby(['make', 'category'], as_index=False)['sell_price'].mean()

# Create separate plots for each category
plt.figure(figsize=(25, 20))

# Set the style for the plots
sns.set_style("whitegrid")

# Order of categories
ordered_categories = ['High', 'Medium', 'Economical']

# Loop through categories to plot each
for i, category in enumerate(ordered_categories):
    plt.subplot(3, 1, i + 1)  # Create 3 vertical subplots
    # Filter the average prices for the current category and sort them
    category_data = df_avg[df_avg['category'] == category]
    # Use a softer color palette
    palette = sns.color_palette("pastel")
    sns.barplot(data=category_data.sort_values(by='sell_price', ascending=False), 
                x='make', 
                y='sell_price', 
                palette=palette)
    
    plt.title(f'Average Selling Price - {category}', fontsize=24, fontweight='bold', color='navy')
    plt.xlabel('Make', fontsize=20)
    plt.ylabel('Average Selling Price ($)', fontsize=20)
    plt.xticks(rotation=45, ha='right', fontsize=16)
    plt.yticks(fontsize=16)
    plt.grid(axis='y', linestyle='--', alpha=0.7)
    
    # Add value labels on top of bars
    for p in plt.gca().patches:
        plt.annotate(f'{p.get_height():,.0f}', (p.get_x() + p.get_width() / 2., p.get_height()),
                     ha='center', va='bottom', fontsize=12, color='black')

plt.tight_layout()  
plt.show()

plt.figure(figsize=(10, 6))
sns.boxplot(data=df, x='category', y='sell_price', palette='pastel')

plt.title('Box Plot of Selling Prices by Category', fontsize=18)
plt.xlabel('Category', fontsize=14)
plt.ylabel('Selling Price ($)', fontsize=14)
plt.xticks(fontsize=12)
plt.yticks(fontsize=12)
plt.grid(axis='y', linestyle='--', alpha=0.7)

plt.tight_layout()
plt.show()

Category Breakdown:

High Category: Features luxury brands like BMW and Mercedes, appealing to affluent consumers seeking prestige.

Medium Category: Targets middle-class buyers with reliable and efficient options.

Economical Category: Attracts budget-conscious consumers, often including compact and older models

Which states are the most sales-making

state_sales = df.groupby('state')['sell_price'].sum().sort_values(ascending=False)
top_6_states = state_sales.nlargest(5)
other_sales = state_sales.iloc[5:].sum()
other_series = pd.Series({'Other': other_sales})
top_6_states = pd.concat([top_6_states, other_series])

plt.figure(figsize=(12, 12))
colors = sns.color_palette("Set2", len(top_6_states))

explode = [0.05] + [0.02] * (len(top_6_states) - 1)

plt.pie(top_6_states.values,
        labels=top_6_states.index,
        autopct='%1.1f%%',
        explode=explode,
        colors=colors,
        startangle=290,
        wedgeprops={'edgecolor': 'black'},
        textprops={'fontsize': 14})

plt.title('Percentage of Sales by State (Top 5 + Other)', fontsize=18)
plt.tight_layout()
plt.show()

The highest total sales of cars are in Florida, California, Pennsylvania, Texas, Georgia, in conjunction with the number of sales

Which brands have the most models?

brand_car = df.groupby('make', as_index=False)['model'].count().sort_values('model', ascending=False)
top_10 = brand_car.head(4)
other_total = brand_car.iloc[10:]['model'].sum()
other_data = pd.DataFrame({'make': ['Other'], 'model': [other_total]})
combin_data = pd.concat([top_10, other_data])
plt.figure(figsize=(12, 8), facecolor='lavender')
plt.pie(combin_data['model'], labels=combin_data['make'], autopct='%1.2f%%',startangle=270, colors=sns.color_palette('viridis', len(combin_data)))
plt.title('Brand By Models Count', fontsize=20)
plt.axis('equal')
plt.show()

Ford is the car that has a larger number of models, then Chevrolet

Who are the most profitable sellers?

top_sellers = df.groupby('seller')['sell_price'].sum().nlargest(5).reset_index()
top_sellers

	seller	sell_price
0	nissan infiniti lt	456580011.0
1	ford motor credit company llc	306871800.0
2	the hertz corporation	232114601.0
3	avis corporation	183512130.0
4	santander consumer	115271200.0

plt.figure(figsize=(12, 8))
sns.barplot(data=top_sellers, x='sell_price', y='seller', palette='viridis')
plt.title('Top 5 Sellers by total selling price', fontsize=16)
plt.xlabel('Total Sales ($)', fontsize=14)
plt.ylabel('Seller', fontsize=14)
plt.show()

Insight on New Selling Prices by Sellers

1. Price Analysis:

   The data shows that Nissan Infiniti LT tops the list with a new selling price of $456,580,011. This could indicate that this brand's cars are either highly popular or feature luxury models, reflecting a significant market value.

2. Company Comparison:

   Ford Motor Credit Company LLC ranks second with a price of $306,871,800, followed by The Hertz Corporation at $232,114,601. This highlights the influence of well-established brands and their impact on the car sales market.

What is most profitable Body?

avg_body_prices = df.groupby('body')['sell_price'].mean().reset_index()
avg_body_prices = avg_body_prices.sort_values(by='sell_price', ascending=False)  

plt.figure(figsize=(15, 8))
sns.barplot(data=avg_body_prices, x='body', y='sell_price', palette='coolwarm')

plt.title('Average Selling Price by Body Type', fontsize=18)
plt.xlabel('Body Type', fontsize=14)
plt.ylabel('Average Selling Price ($)', fontsize=14)

plt.xticks(rotation=90, ha='right', fontsize=12)

plt.tight_layout()

plt.show()

Multivariant Analysis

Top 3 sales models in the top 6 brands sales

total_sales = df.groupby(['make', 'model']).agg({'sell_price': 'sum'}).reset_index()
top_makes = total_sales.groupby('make')['sell_price'].sum().nlargest(6).index
top_models = total_sales[total_sales['make'].isin(top_makes)]
top_models = top_models.loc[top_models.groupby('make')['sell_price'].nlargest(3).reset_index(level=0, drop=True).index]

other_models = total_sales[total_sales['make'].isin(top_makes) & ~total_sales['model'].isin(top_models['model'])]
other_sales = other_models.groupby('make').agg({'sell_price': 'sum'}).reset_index()
other_sales['model'] = 'Other'

combined_top_models = pd.concat([top_models, other_sales], ignore_index=True)

plt.figure(figsize=(14, 12))

for i, make in enumerate(combined_top_models['make'].unique(), start=1):
    plt.subplot(3, 2, i)
    colors = plt.cm.get_cmap('Set3').colors
    data = combined_top_models[combined_top_models['make'] == make]
    sizes = data['sell_price']
    labels = data['model']

    explode = [0.05 if label != 'Other' else 0 for label in labels]


    plt.pie(sizes,
            labels=labels,
            autopct='%1.1f%%',
            startangle=90,
            colors=colors[:len(sizes)],
            explode=explode,
            shadow=False)


    plt.title(f'Top Models for {make}', fontsize=14, fontweight='bold')
plt.tight_layout(pad=2)
plt.show()

Make vs. Transmission vs. Selling Price by Condition

subset_makes = df['make'].value_counts().index[:4]
df_subset = df[df['make'].isin(subset_makes)]  

g = sns.catplot(x='transmission', y='sell_price', col='make',
                data=df_subset, kind='bar', col_wrap=2, height=4, aspect=1.2, hue='condition_group')

g.set_titles('{col_name}')
g.set_axis_labels('Transmission', 'New Selling Price')
g.fig.suptitle('Make vs. Transmission vs. Selling Price by Condition', y=1.05)

g._legend.remove()

g.add_legend(title='Condition Group', loc='upper right')

plt.tight_layout()
plt.show()

top_makes = df.groupby('make')['sell_price'].sum().nlargest(10).index
filtered_df = df[df['make'].isin(top_makes)]

plt.figure(figsize=(16, 10))

sns.barplot(data=filtered_df, y='make', x='sell_price', hue='transmission', ci=None, dodge=True)

plt.xticks(rotation=45)
plt.title('Top 10 Makes vs. Transmission vs. Total Selling Price')
plt.ylabel('Make')
plt.xlabel('Selling Price')
plt.legend(title='Transmission')

plt.tight_layout()
plt.show()

top_10_expensive_cars_full_details = df[['make','model', 'year', 'condition', 'sell_price']].sort_values(by='sell_price', ascending=False).head(10)


# Display the detailed top 10 most expensive cars table
display(top_10_expensive_cars_full_details)

	make	model	year	condition	sell_price
254862	FORD	F-150	2014	5	37700.0
180046	DODGE	CHARGER	2013	4	37600.0
153096	GMC	SIERRA 2500HD	2011	5	37600.0
478234	FORD	MUSTANG	2012	5	37500.0
261458	FORD	EXPLORER	2014	5	37500.0
87518	BMW	5 SERIES	2012	4	37500.0
107973	BMW	3 SERIES	2014	5	37500.0
148170	LEXUS	GX 460	2012	4	37250.0
386407	CHEVROLET	SILVERADO 2500HD	2012	4	37250.0
16045	FORD	F-150	2014	4	37250.0

Top 10 Car Soldout count

df['full_model'] = df['make'] + ' ' + df['model'] + ' ' + df['sale_year'].astype(str)
model_n = df.groupby(['full_model', 'condition']).size().reset_index(name='count')

top_models = model_n.groupby(['full_model'])['count'].sum().reset_index()
top_models = top_models.sort_values('count', ascending=False).head(10)
plt.figure(figsize=(16, 8))

sns.barplot(data=model_n[model_n['full_model'].isin(top_models['full_model'])], x='full_model', y='count', hue='condition')

plt.xlabel('Model Name', fontsize=15)
plt.ylabel('Number Of Cars', fontsize=15)
plt.title('Top 10 Car Soldout count', fontsize=26)
plt.xticks(rotation=90, fontsize=14)
plt.yticks(fontsize=14)

plt.legend(title='Condition', fontsize=12)
plt.show()

3D Plot: Year vs Condition vs Selling Price

fig = plt.figure(figsize=(12, 8))
ax = fig.add_subplot(111, projection='3d')
sc = ax.scatter(df['year'], df['condition'], df['sell_price'],
                c=df['sell_price'], cmap='viridis', s=50, alpha=0.7)

ax.set_xlabel('Year', fontsize=14)
ax.set_ylabel('Condition (1-5)', fontsize=14)
ax.set_zlabel('Selling Price', fontsize=14)
ax.set_title('3D Plot: Year vs Condition vs Selling Price', fontsize=16)

cbar = plt.colorbar(sc, ax=ax, shrink=0.6)
cbar.set_label('Selling Price')

plt.tight_layout()
plt.show()

3D Plot: Condition Group vs. Odometer vs. New Selling Price

# Create a 3D scatter plot for Condition Group vs. Odometer vs. New Selling Price
fig = plt.figure(figsize=(10, 7))
ax = fig.add_subplot(111, projection='3d')

# Scatter plot with color based on new_selling_price
scatter = ax.scatter(df['condition'], df['odometer'], df['sell_price'],
                     c=df['sell_price'], cmap='coolwarm', s=50)

# Setting labels
ax.set_xlabel('Condition Group')
ax.set_ylabel('Odometer')
ax.set_zlabel('Selling Price')
ax.set_title('Condition Group vs. Odometer vs. Selling Price')

# Add color bar for reference
cbar = fig.colorbar(scatter, ax=ax, pad=0.1)
cbar.set_label('Selling Price')

plt.show()

3D Plot: Year vs. Odometer vs. Selling Price

fig = plt.figure(figsize=(10, 7))
ax = fig.add_subplot(111, projection='3d')

sc = ax.scatter(df['year'], df['odometer'], df['sell_price'],
                c=df['sell_price'], cmap='viridis', s=50)

ax.set_xlabel('Year')
ax.set_ylabel('Odometer')
ax.set_zlabel('Selling Price')
ax.set_title('Year vs. Odometer vs. Selling Price')

cbar = plt.colorbar(sc)
cbar.set_label('Selling Price')

plt.show()

plt.figure(figsize=(10, 6))
correlation_matrix = df[['condition', 'odometer', 'mmr', 'sell_price']].corr()
plt.show()

<Figure size 1000x600 with 0 Axes>

sns.heatmap(correlation_matrix, annot=True, cmap='coolwarm', linewidths=0.5)
plt.title('Correlation Heatmap of Key Variables')
plt.show()